strstr

(PHP 4, PHP 5, PHP 7, PHP 8)

strstr — Trouve la première occurrence dans une chaîne

Description

strstr(string $haystack, string $needle, bool $before_needle = false): string|false

Retourne une sous-chaîne de haystack, allant de la première occurrence de needle (incluse) jusqu'à la fin de la chaîne.

Note:
strstr() est sensible à la casse. Pour une fonctionnalité identique, mais insensible à la casse, reportez-vous à stristr().

Note:
Si vous voulez juste déterminer si un needle particulier se trouve dans la chaîne haystack, utilisez la fonction strpos() qui est plus rapide et qui consomme moins de ressources.

Liste de paramètres

haystack

La chaîne d'entrée.

needle

La chaîne à rechercher.

Antérieur à PHP 8.0.0, si needle n'est pas une chaîne de caractères, elle est convertie en un entier et appliqué en tant que valeur ordinal d'un caractère. Ce comportement est obsolète à partir de PHP 7.3.0, et se fier à celui-ci est fortement déconseillé. En fonction du comportement attendu, needle doit être transtypé explicitement en une chaîne de caractère, ou un appel explicite à chr() doit être exécuté.

before_needle

Si vaut true, strstr() retourne la partie de haystack avant la première occurrence de needle (needle étant exclus).

Valeurs de retour

Retourne la portion de la chaîne, ou false si needle n'est pas trouvé.

Historique

Version	Description
8.0.0	`needle` accepte désormais une chaîne vide.
8.0.0	Passing an entier as `needle` is no longer supported.
7.3.0	Passer un entier comme `before_needle` a été rendu obsolète.

Exemples

Exemple #1 Exemple avec strstr()

<?php
$email  = 'name@example.com';
$domain = strstr($email, '@');
echo $domain; // Affiche : @example.com

$user = strstr($email, '@', true);
echo $user; // Affiche : name
?>

Voir aussi

stristr() - Version insensible à la casse de strstr
strrchr() - Trouve la dernière occurrence d'un caractère dans une chaîne
strpos() - Cherche la position de la première occurrence dans une chaîne
strpbrk() - Recherche un ensemble de caractères dans une chaîne de caractères
preg_match() - Effectue une recherche de correspondance avec une expression rationnelle standard

add a note

User Contributed Notes 10 notes

down

laszlo dot heredy at gmail dot com ¶

11 years ago


strstr() is not a way to avoid type-checking with strpos().

If $needle is the last character in $haystack, and testing $needle as a boolean by itself would evaluate to false, then testing strstr() as a boolean will evaluate to false (because, if successful, strstr() returns the first occurrence of $needle along with the rest of $haystack).

<?php
findZero('01234');  // found a zero
findZero('43210');  // did not find a zero
findZero('0');      // did not find a zero
findZero('00');     // found a zero
findZero('000');    // found a zero
findZero('10');     // did not find a zero
findZero('100');    // found a zero

function findZero($numberString) {
    if (strstr($numberString, '0')) {
        echo 'found a zero';
    } else {
        echo 'did not find a zero';
    }
}
?>

Also, strstr() is far more memory-intensive than strpos(), especially with longer strings as your $haystack, so if you are not interested in the substring that strstr() returns, you shouldn't be using it anyway. 

There is no PHP function just to check only _if_ $needle occurs in $haystack; strpos() tells you if it _doesn't_ by returning false, but, if it does occur, it tells you _where_ it occurs as an integer, which is 0 (zero) if $needle is the first part of $haystack, which is why testing if (strpos($needle, $haystack)===false) is the only way to know for sure if $needle is not part of $haystack.

My advice is to start loving type checking immediately, and to familiarize yourself with the return value of the functions you are using.

Cheers.

down

xslidian at lidian dot info ¶

11 years ago


For those in need of the last occurrence of a string:

<?php
function strrstr($h, $n, $before = false) {
    $rpos = strrpos($h, $n);
    if($rpos === false) return false;
    if($before == false) return substr($h, $rpos);
    else return substr($h, 0, $rpos);
}
?>

down

Gevorg Melkumyan ¶

3 years ago


Don't  confuse this function with strtr ) I lost like 1 hour on that

down

gruessle at gmail dot com ¶

13 years ago


Been using this for years:

<?php
/**
*
* @author : Dennis T Kaplan
*
* @version : 1.0
* Date : June 17, 2007
* Function : reverse strstr()
* Purpose : Returns part of haystack string from start to the first occurrence of needle
* $haystack = 'this/that/whatever';
* $result = rstrstr($haystack, '/')
* $result == this
*
* @access public
* @param string $haystack, string $needle
* @return string
**/

function rstrstr($haystack,$needle)
    {
        return substr($haystack, 0,strpos($haystack, $needle));
    }
?>

You could change it to:
rstrstr ( string $haystack , mixed $needle [, int $start] )
<?php

function rstrstr($haystack,$needle, $start=0)
    {
        return substr($haystack, $start,strpos($haystack, $needle));
    }

?>

down

w3b_monk3y at yahoo dot com ¶

15 years ago


If you want to emulate strstr's new before_needle parameter pre 5.3 strtok is faster than using strpos to find the needle and cutting with substr. The amount of difference varies with string size but strtok is always faster.

down

brett dot jr dot alton at gmail dot com ¶

17 years ago


For the needle_before (first occurance) parameter when using PHP 5.x or less, try:

<?php
$haystack = 'php-homepage-20071125.png';
$needle = '-';
$result = substr($haystack, 0, strpos($haystack, $needle)); // $result = php
?>

down

Julian Egelstaff ¶

1 year ago


Lookout for logic inversion in old code!



In PHP 8, if the needle is an empty string, this function will return 0 (not false), implying the first character of the string matches the needle. Before PHP 8, it would return false when the needle is an empty string.



There other string functions that are affected by similar issues in PHP 8: strpos(), strrpos(), stripos(), strripos(), strchr(), strrchr(), stristr(), and this function, strstr()



If you are checking if the return value === false then you will be misled by this new behaviour. You also need to check if the needle was an empty string. Basically, something like this:



<?php

$result = $needle ? strstr($haystack, $needle) : false;

?>

down

-2

trent dot renshaw at objectst dot com dot au ¶

9 years ago


> root at mantoru dot de

PHP makes this easy for you. When working with domain portion of email addresses, simply pass the return of strstr() to substr() and start at 1:

substr(strstr($haystack, '@'), 1);

down

-20

root at mantoru dot de ¶

17 years ago


Please note that $needle is included in the return string, as shown in the example above. This ist not always desired behavior, _especially_ in the mentioned example. Use this if you want everything AFTER $needle.

<?php
function strstr_after($haystack, $needle, $case_insensitive = false) {
    $strpos = ($case_insensitive) ? 'stripos' : 'strpos';
    $pos = $strpos($haystack, $needle);
    if (is_int($pos)) {
        return substr($haystack, $pos + strlen($needle));
    }
    // Most likely false or null
    return $pos;
}

// Example
$email = 'name@example.com';
$domain = strstr_after($email, '@');
echo $domain; // prints example.com
?>

down

-25

leo dot nard at free dot fr ¶

19 years ago


When encoding ASCII strings to HTML size-limited strings, sometimes some HTML special chars were cut.



For example, when encoding "��" to a string of size 10, you would get: "à&a" => the second character is cut.



This function will remove any unterminated HTML special characters from the string...



<?php

function cut_html($string)

{

    $a=$string;



    while ($a = strstr($a, '&'))

    {

        echo "'".$a."'\n";

        $b=strstr($a, ';'); 

        if (!$b)

        {

            echo "couper...\n";

            $nb=strlen($a);

            return substr($string, 0, strlen($string)-$nb);

        }

        $a=substr($a,1,strlen($a)-1);

    }

    return $string;

}

?>

add a note